Klebsiella pneumoniae arbapenemase-producing K. pneumoniae (KPC-Kp) has emerged as an important nosocomial pathogen that is associated with high mortality and treatment failure rates. Colistin and tigecycline remain active against KPC-Kp strains in vitro, but monotherapy with these agents results in suboptimal outcomes. Recent data suggest that combination therapy is superior to monotherapy, especially when the combination contains a carbapenem. Our laboratories have studied the role of the combination of colistin and doripenem (the formulary carbapenem at our hospital) against KPC-Kp infections. We showed that this regimen results in greater time-kills than other combinations against KPC-Kp isolates in vitro, and more reliably achieves bactericidal activity and synergy. We also showed that the level of resistance to doripenem (as reflected by minimum inhibitory concentrations [MICs]) varies widely among clinical KPC-Kp strains, and this might impact the response to colistin-doripenem combination. The goal of this project is to systematically characterize the levels of resistance of KPC-Kp strains to colisitin and doripenem, and to identify predictors for treatment failure of the colistin-doripenem combination. Our central hypothesis is that high levels of resistance to either drug alone, as evident by MICs and/or the presence of specific resistance mechanisms, are the primary determinants of treatment failure. We will test this hypothesis by pursuing two specific aims. First, we will correlate colistin and doripenem MICs with the presence of key resistance mechanisms against these agents (lipid A modifications for colistin, and the expression levels of blaKPC and ompK35/36 encoding major porins for doripenem). Second, we will correlate treatment outcomes of the colistin-doripenem combination against KPC-Kp infections with MICs of individual drugs and the presence of resistance mechanisms. For this aim, we will review the outcomes of patients with KPC-Kp bacteremia and mice with systemic KPC-Kp infections treated with colistin-doripenem. Our preliminary data and proposal are built upon a unique biorepository of over 200 clinical KPC-Kp strains. To our knowledge, this will be the first study that evaluates the effect of a combination regimen against KPC-Kp strains in the context of mechanism of resistance. If successful, the study will generate data that will provide new insights into the treatment of KPC-Kp infections. In addition, it will lead to future studies that link molecular characterization of highly drug-resistnt strains with the efficacy of antimicrobial combinations.